Content uploaded by Bettina Beer
Author content
All content in this area was uploaded by Bettina Beer on Apr 06, 2015
Content may be subject to copyright.
Social Reproduction and Ethnic Boundaries:
Marriage Patterns Through Time and Space
Among the Wampar, Papua New Guinea
By Bettina Beer* and Julia H. Schroedter**
Abstract
In this article we examine how partner choice and strategies of social repro-
duction among the Wampar of northeastern Papua New Guinea are implicated
in currently pressing questions about the future of Wampar as a socio-cultural
unit. We use long-term qualitative and quantitative data based on fieldwork and
census surveys conducted between 1954 and 2013 from the village of Gabsong-
keg to analyse temporal and spatial patterns of partner choice. We are especially
interested in interethnic marriages and their effects on group boundaries and
group identities, given a pre-existing pattern of ethnic endogamy. Our results
show that intermarriages between Wampar and non-Wampar have constantly
been rising; in younger marriage cohorts some 60% of Wampar individuals are
intermarried with partners of other ethnic identities. The data reveal that local
and historical particularities inflect partner choices in ways that impact on set-
tlement patterns, modes of engagement with the economic institutions of the
modern state and, ultimately, the taken-for-granted nature of the identity inher-
ing in the name “Wampar”; these impacts, in turn, increase the likelihood of in-
terethnic marriage and precipitate questions about the rights attaching to local
corporate identities under conditions where land is increasingly related to its
commodity values.
Keywords: intermarriage, partner choice, social relations, Papua New Guinea,
Wampar
1. Introduction
Visitors travelling to Morobe Province, in the northeast of the state of
Papua New Guinea (PNG), land at Nadzab airport. It is located near the
Highlands Highway about 40 kilometres outside of Lae, the provincial
Sociologus, Volume 64, Issue 1, p. 1 –28
Duncker & Humblot, Berlin
Sociologus 64 (2014) 1
*Ethnologisches Seminar der Universität Luzern, Frohburgstrasse 3, CH 6002
Luzern ·E-Mail: Bettina.Beer@unilu.ch.
** Soziologisches Institut der Universität Zürich, Andreasstrasse 15, CH 8050
Zürich ·E-Mail: schroedter@soziologie.uzh.ch.
capital and second-largest city of PNG. The airport is on land leased
from the village of Gabsongkeg, one of several on the territory of mem-
bers of the population bearing the name “Wampar”1. Today they occu-
py a semi-urban area, with much of the population accustomed to en-
gagement with the market economy. Whether the Wampar will remain
a distinct ethnic group is not only an academic question but also an is-
sue frequently discussed among themselves. Such discussions arise in
diverse contexts initiated by actors in different social positions. One of
the most pressing problems in 2013 was the selling of land to non-
Wampar and conflicts over the rights to sell or lease land and who
should get shares of the money. Several women were in conflicts with
their brothers who, based on “traditional”patrilineal law, sold family
land. The women argued that for modern financial transactions patri-
lineal land rights do not matter and that selling is not allowed anyway.
Other contexts are pre- and elementary schools where Wampar parents
discuss problems (real and assumed) of their kids outnumbered by ya-
ner2. Other areas of discussion of Wampar-ness are questions of access
to health-care and village courts or yaner not following Wampar values
and rules of closing the market on Sundays. Increasing violence, drug
consumption, and the spreading of HIV/ AIDS lead women’s groups,
church activists but also ordinary Wampar to conclude that maintain-
ing an ethnic identity and unit would help against these ‘outside
threats’. The advent of mining and the imagined benefits (employment,
royalties, income from subcontracting) and dangers (environmental
pollution, being overrun by yaner) open up further areas for concern
and discussion of ethnic unity in which nearly every Wampar is in-
volved.
In times when their own language is used less and less frequently (as
PNG’s lingua franca, Tok Pisin, grows in importance), individuals sell
2 Bettina Beer and Julia H. Schroedter
Sociologus 64 (2014) 1
1When we write about “the Wampar”, the reader should keep in mind that
“Wampar”(like other ethnic groups in PNG such as the Maring, LiPuma 2000)
have not always been a bounded social unit with its own territory. In pre-colonial
times various groups fought and moved out of the mountains to settle the Mark-
ham Valley and coalesce into a population bearing the ethnonym “Wampar”. An
overarching ethnic identity was strengthened or even substantially produced by
missionaries as well as other colonial and post-colonial transformations. As in
other parts of Papua New Guinea “ethnicity is based on continua of cultural dif-
ference in a population crisscrossed by flows of people”(Golub 2014: 118) and not
on clearly bounded entities.
2Wampar call all non-Wampar from PNG yaner, they address and refer to spe-
cific people mostly with their first name and a place name of their area of origin,
e.g. “Kaspar Sepik”. The ethnonyms we use in this article are used by Wampar for
a specific person and refer to localities of different sizes, ethnic groups (“Simbu”)
and regions (“Hailans”). Sometimes they match the ethnonyms used by yaner
themselves.
land to non-Wampar (called ‘yaner’), and many aspects of routine social
interaction are transformed by the prospects of large-scale mining
(Bacalzo, Beer and Schwörer in press), the loss of their ethnic identity
has become a real fear for many Wampar.
Figure 1: Map of PNG and Wampar Villages (Map H. Schnoor)
This article examines how patterns of marriage and social reproduc-
tion3contribute to the transformations of “Wampar-ness”and the fu-
ture of this population as a socio-cultural unit. We will analyse tempor-
al and spatial patterns of partner choice and marriage in the village of
Gabsongkeg, focusing on interethnic marriages between Wampar and
yaner, as well as the partner choices of the descendants of these inter-
marriages. Interactions between Wampar and non-Wampar have a long
history, beginning with inter-ethnic exchange and warfare before the
arrival of Lutheran missionaries in 1909 and the movement of Wampar
evangelists into neighbouring Watut communities after pacification in
the 1920s. Current patterns of interethnic marriages across time and
Social Reproduction and Ethnic Boundaries 3
Sociologus 64 (2014) 1
3By using the term ‘social reproduction’we refer to the reproduction of social
structures and systems with an emphasis on “the repetitive character of day-to-
day life, the routines of which are formed in terms of the intersection of the pas-
sing (but continually returning) days and seasons.”(Giddens 2007: Time, the Body,
Encounters) and not to the consolidation of social cohesion as Giddens clarifies.
space seem to imply a specific dynamic in their distribution that we
want to expose. More specifically, we aim to explain the spatial and
temporal processes affecting marriage patterns in terms of local and
more general factors conditioning social reproduction.
Our analysis is in the spirit of Anthony Giddens’s (2007) theory of
structuration in linking the temporal and spatial positioning of actors
to the possibility of the reproduction or transformation of Wampar so-
cial relations as a system. Giddens insists that structures are produced
and reproduced constantly by the actions of the individuals, even while
social resources (from rules to institutions) are a condition of individual
action. Under the contemporary circumstances obtaining among the
Wampar, in PNG, access to economic values associated with land are
still tied up with ethnic and local corporate identities and the politics
of affinity, notwithstanding the burgeoning pressures to bring land
within the market economy as a means of non-traditional production.
Our account is an analysis of the effects at an aggregate level of the
exercise of agency –referring to the capability of persons to take inten-
tional action (Giddens 2007) –in the context of marriage and the chang-
ing historical profile of ethnic endogamy among the Wampar. Given
the crucial role of marriage in social reproduction, and its explicitly ac-
knowledged role in relations between corporate entities implicated in
patterns of land use, marriage choices are especially liable to have
trans-generational effects that are radically transformative of local
structural relations. The situatedness, in time and space, of Wampar
and non-Wampar partners (and potential partners) leads to patterns of
partner choice which can only be explained by their subjectivity as it is
inflected by historically situated constraints and chances. These con-
straints and opportunities have changed over recent generations; also,
however, the constraints and marital opportunities have resulted in
unions that have reconfigured local fields of social relations in ways
that feedback (negatively or positively) on subsequent choices. It is
these we wish to underline in this paper.
2. The Wampar: Historical and Sociocultural Context
of Interethnic Marriages
The Wampar are one of the several hundred ethnically identified peo-
ples of PNG. Such identities remain relevant to the distribution of poli-
tico-economic resources in times of increasing urbanization and the
expansion of extractive industries. The state itself recognises “tradi-
tional”ethnic identity as relevant to access to land and other economic
resources, but the legacy of PNG’s colonial regime continues to find ex-
4 Bettina Beer and Julia H. Schroedter
Sociologus 64 (2014) 1
pression in the manner in which ethnic identity remains an aspect of
the distribution and structure of social fields (education, physical and
social mobility and local distributions of power).
The Wampar (others called them “Laewomba”) were first mentioned
in reports of German gold miners and colonial officers at the beginning
of the 20th century. After peaceful initial contacts with the medic and
ethnographer, Richard Neuhauss, and missionaries of the “Neuendet-
telsauer Mission”in 1909, a mission station near Gabsongkeg village
was built in 1910/ 11. The Wampar, like other early contacted lowlands
populations in PNG, are relatively well educated (degrees reaching
from high school, over vocational colleges to university) and many men
and women have a stable income from (self-)employed work as teacher,
craftsmen, employees, or graduates (e.g. law, geology, philosophy). In
addition, they are settled on fertile land and had for a long time regular
cash income by selling betel nuts in the markets concentrated in and
around Lae. These regionally significant advantages make their settle-
ment area attractive for migrants, and make the Wampar attractive
marriage partners.
The Wampar live in eight villages. Five of them are situated along the
Highlands Highway, which connects the coastal city of Lae with the
Highlands provinces (see figure 1). In the 1970s, traffic between Lae
and the Highlands increased when the Highlands Highway was up-
graded from gravel to an asphalt road. Airstrips from World War II,
which had been built on Wampar land, were extended and developed to
become Lae’s airport in the 1980s. South of the Markham two villages
are located near the road to Wau and Bulolo. Wampar traditions say
that they came down into the Markham Valley via the Watut, a river
that drains the mountainous interior to the south, under pressure from
expanding populations in the higher altitudes (Fischer 1992: 16, Fischer
2013). They probably drove away earlier inhabitants in claiming their
current territory.
Many Wampar families still depend upon subsistence production for
the bulk of their food: they maintain gardens that supply bananas, their
staple, and coconuts, vegetables, corn, onions, tomatoes, pineapples,
watermelons, taro, yams, sweet potatoes, and peanuts. Areca palms, the
source of the mild narcotic, betelnut, which is chewed all over PNG,
were planted for consumption and for the market. A recent disease of
the areca palm meant that cocoa became an important cash crop. Rice,
sugar, tea, bread, biscuits, and canned goods (sardines and corned beef)
that are mostly purchased from small tradestores supplement garden
produce. Today, overall, the market economy is almost as important as
the subsistence economy, for there are many necessities that can only
be obtained with cash. Many Wampar work in towns, at the airport, or
Social Reproduction and Ethnic Boundaries 5
Sociologus 64 (2014) 1
as teachers, nurses and mechanics, or are engaged in commercial activ-
ities.
After the arrival of Lutheran missionaries and the pacification of the
Markham Valley, the Wampar population began to increase. A 1937
census listed 1,841 Wampar (Vial 1937: 384); the national census of
1980 enumerated 5,150. Intermarriage between Wampar and yaner be-
gan to increase soon after pacification. Initially, the evidence suggests,
intermarriage was predominantly between Wampar men and and yaner
women, but after the Highlands Highway and Lae’s airport were built,
local settlement and marriage patterns have changed dramatically. The
2000 national census, which does not differentiate according to ethnic
group or language (Paliwala 2012: 6), gives a total number of 2,517 citi-
zens for the ward of Gabsongkeg. Official estimates for the year 2000
suggest that the Wampar population stands at approximately ten thou-
sand people, a fivefold increase in sixty-three years (National Statisti-
cal Office 2001). Nevertheless, demographic research indicates that
Wampar fertility is declining (Kramp 1999), possibly as a consequence
of Wampar women’s interest in family planning, which many say is due
to fear that their land will not be sufficient to support coming genera-
tions.4Whatever factors underlie declining fertility, it is clear that the
initial rise of the population was mainly caused by immigration.
Formerly, a Wampar person (ngaeng Wampar) was defined by birth,
socialization, and affiliation to a named clan (which factors rarely came
apart), in contrast to outsiders, those born and raised elsewhere. As
elsewhere in Melanesia, the system always allowed for “adoption”into
a kin group, but the individual’s clan identity and association with spe-
cific places was rarely unclear. Over recent years, intermarriage and
new questions of affiliation have made it more difficult to construct
clear categories differentiating Wampar and outsiders. Contestation
over descent-group and ethnic identity has become stronger with in-
creased prospects for mining (which entails differential employment
and royalty entitlements), the expansion of the Highlands Highway to a
four-lane road (entailing compensation payments and loss of residen-
tial plots) and the growth of many associated economic activities in the
Markham Valley. How international capital articulates with local in-
equalities is a current research project of the first author and collea-
gues. The following section will provide information on the fieldwork
and the data used in the analyses.
6 Bettina Beer and Julia H. Schroedter
Sociologus 64 (2014) 1
4Other factors might be the level of education and infertility due to the spread
of veneral diseases (Beer 2008: 99–100; Kramp 1999: 364), intermarriage per se
seems not to have direct influence as the number of children in mixed families is as
diverse as in Wampar-Wampar families.
3. Methods, Sources and Data
Our analyses are based on both quantitative and micro-focused qua-
litative data. In our view, neither data-set would alone be sufficient to
understand geographical and temporal aspects of the transformation of
Wampar social relations. The data is mainly based on anthropological
fieldwork that Bettina Beer has conducted in the village of Gabsong-
keg. She did fieldwork in Gabsongkeg in 1997, 1999/2000, 2002, 2003 /
04, 2009, and in 2013. This fieldwork was additionally informed by the
Research Focus Wampar, established by Hans Fischer in 1958. Fischer
had conducted fieldwork in Gabmadzung in 1965, and then in Gab-
songkeg in 1971/ 72, 1976, 1988, 1990, 1993, 1997, 1999/ 2000, 2003/ 04
and 2009. In 2009/ 2010, Doris Bacalzo and Tobias Schwörer did re-
search in Dzifasing, and Heide Lienert, Christiana Lütkes, Rita Kramp,
and Juliane Neuhaus worked in different Wampar villages (Bacalzo,
Beer and Schwörer, in press; Fischer 1975, [ed.] 1978). Beer’s data is
complemented by information on marriages5from a ‘village register’
(1954) instituted by the colonial Australian administration, and from
earlier ethnographic censuses, conducted by Hans Fischer in 1971, 1993,
and 1997 (Fischer 1997). In 1999/ 2000, 2003/ 04 and 2009 Beer con-
ducted an ethnographic census in Gabsongkeg that includes stand-
ardized basic data on all households, each member in residence at the
time of the visit (de facto population), and those who are attached but
absent (de jure population). Basic data includes sex, age, civil status,
education, residence, group affiliation (ethnic group, descent-group),
kin relations between members, and location of the household, but in-
cludes less standardized information on migration, education or any to-
pic of specific interest to the ethnographer (see Schulze, Fischer, Lang
1997). All persons living in Gabsongkeg, as defined by the official vil-
lage boundaries, including all household members judged to be only
temporarily absent were listed and described. Beer also conducted many
biographical interviews with one or both partners to interethnic mar-
riages in 2003/ 04, and again in 2009, and had many open, unstructured
conversations with couples and their kin, neighbours, and friends.
In total, we have information on 958 marriages, including marriages
ended by separation or death (n=366) in addition to still-existing (n=592)
Social Reproduction and Ethnic Boundaries 7
Sociologus 64 (2014) 1
5A‘marriage’is marked by the woman’s moving into her partner’s household:
the term itself refers to the cohabitation of the partners regardless of whether or
not they underwent a church ritual, which –if it happens at all –often occurs much
later, or their union is legally registered (a very rare event). The definition implies
a certain amount of imprecision, as partners might move in together and then se-
parate again, for various reasons and for indeterminate lengths of time. The exact
status of a relationship might be contested after a partner leaves to stay with par-
ents after a conflict, with some people but not others construing it as a divorce.
marriages. The total comprises interethnic (n=538) and intraethnic
(n=420) marriages involving at least one partner from Gabsongkeg
village; the earliest documented couples were married in the 1890s, the
latest in 2009. The Gabsongkeg data is not representative for the Wam-
par area as a whole: villages differ in distance to the highway and the
town of Lae, in the sharing of boundaries with other ethnic groups and
in employment possibilities.
We are confident that Beer’s long-term ethnographic fieldwork has
yielded good, reliable qualitative data (Beer 2006a, 2006b, 2008, 2010).
However, the ethnographic census data poses some challenges for the
quantitative analysis. As the data was not collected in a fully stand-
ardized manner, some variables suffer from a great number of missing
values. Investigations in societies without official calendars and insti-
tutionalized registration procedures must rely on estimates and indirect
inferences to yield dates and related data. Accordingly, more fine-
grained multivariate analyses cannot be conducted. But the descriptive
results already allow insights into the marriage patterns of the Wampar
that can guide and enrich the qualitative data analysis.
4. Ethnic Origin of Partners,
Socioeconomic Status and Gender
Figure 2 presents the trends of intermarriages between Wampar and
non-Wampar (yaner) over the marriage cohorts from 1920 to 2007 (pre-
sented by a five-year moving average) differentiated by sex.
Figure 2 includes all marriages of individuals (“WamparI”) generally
accepted as being Gabsongek residents of Wampar origin, which means
that we have included those with one Wampar parent, and who, ac-
cordingly, are also the offspring of intermarriages.6Moreover, we fol-
low Wampar practice in including persons who became Wampar by
adoption.
The graphs show an increase over the whole time period for Wampar
men and women alike. However, intermarriages between Wampar men
and non-Wampar women (afi yaner) began earlier –mainly in the
1930s, when missionary powers were at their height –than between
Wampar women and non-Wampar men (ngaeng yaner), which first oc-
curred after the Second World War. Overall, the difference between the
8 Bettina Beer and Julia H. Schroedter
Sociologus 64 (2014) 1
6We refer to these individuals as “WamparI”(for inclusive), which embraces
mixed Wampar (“WamparM”) and Wampar only “WamparO”with both parents
being of Wampar origin. In section 3 and 5 we differentiate WamparM with a ya-
ner father from those with a yaner mother.
Source: Ethnographic census data 1954 to 2009, n=670 (men); n=620 (women).
Note: A marriage between a Wampar and a Wampar / Adzera is not counted
as intermarriage but as intraethnic marriage.
Figure 2: Trends of Intermarriages of WamparI Men and Women
over Marriage Cohorts (Five-year Moving Average), in Percentages
genders is not stable over the marriage cohorts, but Wampar men are
slightly more likely to marry someone outside their ethnic group than
Wampar women; in the most recent marriage cohorts this tendency is
more marked. By 2009, some 60–70% of all Wampar marriages had
been to a yaner.
The increase of the Wampar population in the 1980s and 1990s was
caused mainly by migration from the Watut River, the Highlands, the
Sepik Provinces, and the neighbouring Erap and Adzera areas to the
Markham Valley and Lae. Desire for access to schools, hospitals, jobs,
cash and the “modern world”induced people to leave their hinterland
villages and come to coastal towns. Wampar became more mobile too.
Increasing migration and mobility resulted in more marriages with
yaner which increased the opportunities for others to visit Wampar set-
tlements to stay with inmarried relatives and the chances of further
marriages. This section analyses Wampar-yaner marriages in terms of
gender, affiliation to ethnic groups and their economic status, which re-
veals aspects of exchange relations between kin-based networks and
ethnic groups.
Social Reproduction and Ethnic Boundaries 9
Sociologus 64 (2014) 1
As described above, in the calculation of the trends of intermarriages
we did not count descendants of interethnic couples resident on Wam-
par territory as yaner because, by default, Wampar designate them as
“Wampar”. Children of yaner fathers are more often excluded on the
basis of patrilineal considerations, but practice is not consistent (Bacal-
zo 2012). By counting the offspring of intermarriages as Wampar, the
percentage of Wampar-Wampar marriages looks higher than it other-
wise might, but doing so reflects how Wampar categorize people in the
majority of cases. This also explains much of the overall growth of the
Wampar population. We will include offspring of interethnic couples
also in the statistics on personal characteristics of the marriage part-
ners (next section) but will analyse their partner-choice separately
when it comes to kinship networks, intermarriages and the future of
Wampar social boundaries.
It is not possible to get information on the ethnic origin or language
group of migrants on Wampar territory from the official PNG Nation-
al Census of 2000 and 2010 (National Statistical Office 2001, 2013).
Hence, collecting data on the ethnic origin of in-marrying partners has
been an important task in the ethnographic census although it is de-
manding: Wampar –even close relatives of an in-married person –give
names for whole areas (hailans ‘from the mountains’or nambis ‘from
the coast’), PNG provinces, well known ethnonyms (Tolai), or towns lo-
cated in the area from which a person comes (Kerema). Several of these
categories can overlap. Bettina Beer talked to some of the migrants
themselves and then listed the larger category, the most usual ethno-
nym, and smaller units down to the specific village / place name. The re-
sult was about 50 labels for the origin of yaner which we reduced for
our analysis to categories according to spatial distance and socio-eco-
nomic hierarchies: ‘Direct neighbours’and ‘Fringe peoples’(further
away from town and economically disadvantaged compared to Wam-
par), ‘Highlands’(a large category of individuals characterized by the
following features: further away, in the mountains of the central cordil-
lera of New Guinea, negatively stereotyped among Wampar, very
mixed economic background), ‘Distant socioeconomic equal’and ‘Dis-
tant poorer’groups, which include many coastal and islands popula-
tions. The last category ‘Outside PNG’, with only four marriages, can
be ignored.
Table 1 and table 2 report the origin of the partners of Wampar men
and women respectively. Furthermore, we differentiate between Wam-
parI, WamparO and WamparM, descendants of mixed marriages be-
tween Wampar and other ethnic groups (cf. footnote 5). The majority of
the marriages of Wampar men and women are homogamous in ethnic
terms, i.e. to Wampar spouses. Direct neighbours (Adzera, Erap, Watut,
10 Bettina Beer and Julia H. Schroedter
Sociologus 64 (2014) 1
and Yalu) who share a boundary with the Wampar are also important
marriage partners, especially for Wampar men. Among them, the Ad-
zera, who are the next neighbours up the Highlands Highway, are
clearly the most important source of yaner married by Wampar, with
103 (71 Adzera women and 32 Adzera men) out of the total of 538 inter-
ethnic marriages. The Adzera occupy an economically less advanta-
geous area than the Wampar and intermarriage has a long history. Such
marriages clearly tend to a pattern of virilocal residence and involve
the exchange of bridewealth (a pig, bananas and money) shortly after
marriage, whereas intra-Wampar affines exchange bridewealth much
later.
Table 1
Origin of Female Partners (
Afi Yaner
)
of WamparO and WamparM Men, Percent in Columns
Origin of woman Origin of man Total n
WamparO WamparM WamparI
Other 0.8% 6.5% 1.3% 9
WamparM 4.0% 8.1% 4.3% 31
WamparO 56.1% 37.1% 54.5% 389
Direct Neighbours 13.8% 11.3% 13.6% 97
Fringe peoples 4.1% 8.1% 4.5% 32
Highlands 4.1% 3.2% 4.1% 29
Distant socioeconomic
equal groups 14.0 % 24.2 % 14.8 % 106
Distant poorer groups 2.9% 1.6% 2.8% 20
Outside PNG 0.2% 0.0% 0.1% 1
Total 100.0% 100.0% 100.0% 714
Source: Ethnographic census data 1954 to 2009, n=652 (WamparO); n=62 (WamparM).
Table 2
Origin of Male Partners (
Ngaeng Yaner
)
of WamparO and WamparM Women, Percent in Columns
Origin of man Origin of woman Total n
WamparO WamparM WamparI
WamparM 4.0% 5.9% 4.0% 23
WamparO 63.3% 30.6% 63.3% 392
Direct Neighbours 6.2% 15.3% 6.2% 49
Fringe peoples 0.7% 2.4% 0.7% 6
Highlands 1.7% 8.2% 1.7% 17
Continued next page
Social Reproduction and Ethnic Boundaries 11
Sociologus 64 (2014) 1
Table 2 (continued)
Origin of man Origin of woman Total n
WamparO WamparM WamparI
Distant socioeconomic
equal groups 18.0 % 32.9 % 18.0 % 132
Distant poorer groups 5.7% 3.5% 5.7% 36
Outside PNG 0.3% 1.2% 0.3% 3
Total 100.0% 100.0% 100.0% 663
Source: Ethnographic census data 1954 to 2009, n=578 (WamparO); n=85 (WamparM).
The figures shown in the tables reflect the advantages discussed in
the foregoing paragraph that their economically superior position gives
Wampar men in search of an often younger (see section 5) spouse:
13.6% of all Wampar men marry a spouse from their direct neighbours
while only 6.2% of Wampar women do so. The long-standing nature of
affinal links between Wampar and the Adzera is also reflected in the
place of residence of the couples: many Wampar-Adzera couples live in
the principal Wampar settlements, whereas later in-marrying yaner are
more likely to live in isolation along the Highlands Highway (see ta-
ble 5). In-marrying Adzera men sometimes maintain two households
(an uxorilocal and a virilocal one) or live exclusively in their place of
origin in order not to lose their patrilineally inherited rights to land.
Sharing a territorial boundary, frequent possibilities of contact and in-
teraction (e.g. visits, exchange of bride wealth) as well as already exist-
ing kinship networks facilitate further intermarriages. However, these
factors cannot explain the gender difference in the proportions of inter-
marriage (nor can any demographic factor like an uneven gender ratio
among the Adzera, National Statistical Office 2013: 25): these are more
readily explained in terms of the attractiveness of proximity to urban
opportunities that residence on Wampar territory provides (for more on
the influence of siblings on marital and post-marital residence choices
see Beer in press).
Afi Wampar seem to be more inclined to marry ngaeng yaner from
distant socioeconomically similar groups (18.0%). Indeed, 32.9% of afi
Wampar of mixed parentage (WamparM) marry into these groups. Very
few Wampar women (0.7%) marry men from the remoter, poorer areas
between the Markham Valley and the highlands. These patterns indi-
cate a tendency for Wampar women to avoid marriages to men from so-
cio-economically disadvantaged groups. Case studies and interviews
show that in long-standing marriages, yaner husbands are usually bet-
ter educated than their Wampar wives, or that these men have at least
a comparable economic and educational background.
12 Bettina Beer and Julia H. Schroedter
Sociologus 64 (2014) 1
The tradition of ngaeng Wampar taking an afi yaner as a second or
third wife is consistent with the general pattern of female hypergamy
in interethnic marriages. Polygynous marriages show marked status
differences between the groups of origin of the spouses. Such marriages
are recorded in the early accounts by missionaries (Fischer [ed.] 1978:
96 ff.): Wampar men took women captive during fighting with neigh-
bouring groups or other Wampar villages and brought them home as
wives. Bridewealth was exchanged only for the first wife, who enjoyed
a superior position within the household. Despite all efforts of the mis-
sionaries, polygyny has never been given up completely after pacifica-
tion, Wampar men still took women from other ethnic groups as second
or third wives. In our sample there are no polygynous marriages among
ngaeng yaner or their male offspring.
Today, social relations are more complex, with the result that the
classification of unions is more difficult: there are still straightfor-
wardly polygynous marriages, but with changes of values, patterns of
partner choice, ambiguities in cohabitation and the fragility of mar-
riages, it is not always clear if a marriage is ‘polygynous’, in the earlier
sense of a permanent living arrangement and a clear hierarchy within
the household. Today different relationships with Wampar and / or ya-
ner women can overlap for variable, often indeterminate periods. Thus,
a first marriage might have ended in a divorce while a second relation-
ship remain ambiguous, as between a new marriage and a short-term
love affair or both relationships might co-exist for a period during
which nobody is completely clear about the future. Accordingly, the
number of unclear cases in our sample is quite high (1.8%). Nonethe-
less, 2.4% (n=23) marriages of Wampar men are unequivocally polygy-
nous marriages, in none of which is a Wampar woman the second or
third wife. Increase in intermarriage and contemporary patterns of
polygyny both suggest a regional pattern of hypergamy, with a net
movement of women as spouses into the socio-economically privileged
Wampar community.
Interactions (peaceful and warlike) between Wampar and their
neighbours have existed for a long time, offering structural opportu-
nities to meet marriage partners. Later, the increasing wealth of some
coastal groups, urban New Guineans or other well-off ethnic groups
led to their greater participation in education, increasing social and
spatial mobility, individual affluence and opportunities to travel.
Greater numbers of men have been coming to Lae and travelling
through the settlement area of the Wampar, which offers opportunities
for Wampar women to meet attractive future marriage partners. So,
the specific pattern of social mobility and hypergamy in the Markham
valley occurs in the context of PNG wide patterns of increased mobil-
Social Reproduction and Ethnic Boundaries 13
Sociologus 64 (2014) 1
ity, and social relations in schools, workplaces and towns. We will ana-
lyse the places where couples meet and the ways they get in contact in
section 6 after giving some further information on the traits of the in-
terethnic partners (section 5).
5. Characteristics of and Exchanges
Between Interethnic Partners
First, we focus on the age at marriage of Wampar men and women,
which might shed light on the patterns of intermarriage (cf. table 3).7
Second, we describe the age differences between partners in inter- and
intraethnic marriages. In this section, statistics are again presented for
all Wampar individuals (WamparI), as there are no noteworthy differ-
ences between the offspring of intermarriages and those with two
Wampar parents.
Table 3
Age at Marriage for Male and Female WamparI
in Intra- and Interethnic Marriages, in Years
Age at marriage Minimum Maximum Mean Median
WamparI men
Intraethnic 15 59 24.5* 23
Interethnic 15 62 27.2* 26
Combined 15 62 25.6 24
WamparI women
Intraethnic 14 49 22.2 21
Interethnic 14 55 22.3 21
Combined 14 55 22.2 21
Source: Ethnographic census data 1954 to 2009, n=697 (men); n=592 (women), * p<0.001
(between marriage type within each sex).
The age at marriage of the partners in intermarriages and in Wam-
par-Wampar marriages only differs for Wampar men. They are on aver-
age 24.5 years old when they marry a Wampar wife (i.e. when the cou-
ple moves in together). If they marry an afi yaner, Wampar men are sig-
nificantly older; on average these marriages are entered at the age of
14 Bettina Beer and Julia H. Schroedter
Sociologus 64 (2014) 1
7Older individuals might be more likely to intermarry as they should have fewer
opportunities to find a non-married Wampar spouse.
27.2 years. For Wampar women we do not find statistical significant
differences in respect to the age at marriage. Whether they have an in-
terethnic or an intraethnic marriage, Wampar women are most likely to
marry at the age of 22 years.
The higher age at marriage of Wampar men over women can partly
be explained by the fact that it is often not their first marriage: some of
them marry an afi yaner as a second wife to an already existing rela-
tionship as described above, or after a first marriage ended by separa-
tion or death of their first wife.8Death in connection with pregnancy
and childbirth still happens often and is a general problem in Papua
New Guinea.9Nonetheless, the age difference between partners –al-
though smaller –still remains significant if polygynous marriages are
excluded.
Now we turn to the age difference between partners within inter-
versus intraethnic marriages: many Wampar men marry much younger
yaner wives. Table 4 shows the age difference in intra- and interethnic
marriages, which on average is 2.1 years for Wampar-Wampar mar-
riages. This is consistent with the findings of Walter Schulze’s earlier
analysis of Fischer’s 1971, 1976 and 1988 ethnographic census. Schulze
excluded all interethnic marriages and showed that the data for 61
marriages was in an amazing accordance of reality with the Wampar
ideal, that partners should be nearly of the same age at marriage, the
man being only slightly older than his wife. In most cases the age differ-
ence was 2 to 3 years (Schulze 1997: 97).
Wampar men married to an afi yaner are on average 7.1 years older
than their wives, whereas the average age difference between Wampar
women and their ngaeng yaner husbands is on average 1.8 years and
thereby not significantly different from that of Wampar women in in-
traethnic marriages. Wampar men marry much younger partners from
non-Wampar ethnic groups, which suggests that they are interested in
younger women, who are seen as physically attractive and stronger
than older women. Afi yaner from remote areas who meet a future
Wampar partner might also prefer partners of the same age but could
be willing to marry –in some cases –much older men in the absence of
Social Reproduction and Ethnic Boundaries 15
Sociologus 64 (2014) 1
8Actually, almost 26% of the Wampar men in interethnic marriages have been
married before compared to only 11% of those in intraethnic marriages; for wo-
men the respective percentages are 11% versus 8%. Accordingly, we ran further
analyses in which we only included first marriages. Although the differences be-
tween the types of marriages slightly reduced in part, the overall results (e.g. re-
garding significant differences) remained the same.
9In 2009 a maternal mortality rate (MMR) of 733 deaths in every 100,000 live
births was reported for 2006 by the Department of Health (National Statistics Of-
fice 2009: 109–11).
Table 4
Age Difference Between Partners
in Intra- and Interethnic Marriages, in Years
Age difference between partners
(age of men –age of women) Minimum Maximum Mean Median
WamparI men
Intraethnic –20 26 2.1* 2
Interethnic –3 40 7.1* 5
Combined –20 40 3.2 2
WamparI women
Intraethnic –20 26 2.1 2
Interethnic –6 10 1.8 2
Combined –20 26 2.1 2
Source: Ethnographic census data 1954 to 2009, n=450 (men); n=417 (women), * p<0.001
(between marriage type within each sex).
(younger) alternatives. Living with an older man on Wampar territory
offers them advantages that even compensate for problems associated
with the decision (having no support from kin being far from home,
having to care for an aging husband, being low in the social hierarchy):
proximity to town and booming markets, fertile land for gardening,
availability of education and good health services.10 This pattern can be
interpreted as an exchange of youth and physical attractiveness against
economic benefits and/ or status, which heightens the utility of the
match for each individual. Several studies present empirical evidence
on the exchange between status or economic resources of males and the
attractiveness of females in partner selection (e.g. Schoen and Wool-
dridge 1989, Franzen and Hartmann 2001). It is consistent with the
findings of Buss and his colleagues that men place more emphasis on
the physical attractiveness of potential partners whereas women em-
phasize the socioeconomic resources of potential partners (Buss 1989,
16 Bettina Beer and Julia H. Schroedter
Sociologus 64 (2014) 1
10 Among Wampar physical and sexual attractiveness are discussed for both
sexes referring to health and age. Physical attractiveness is here used in a wider
sense including ability to work in a garden and reproductive success not restricted
to aesthetic beauty ideals. Nevertheless we appreciate that the pattern we are de-
scribing would appeal to evolutionary psychologists committed to the view that
the sexes have different strategies in order to maximize reproductive success.
These involve the exchange of youth and physical attractiveness against economic
standing and/ or social status which are especially relevant for men (Buunk et al.
2001, Grøntvedt and Kennair 2011, Schwarz and Hassebrauck 2012). However a
proper consideration of these perspectives is beyond the purview of this paper so
we merely note this convergence.
Buss and Angleitner 1989, Feingold 1990). This pattern has been found
in various cultures (Buss 1989), although the difference between the
genders seems to have lessened in more recent years (Buss et al. 2001).
Yaner women are also said, by Wampar men, to be less self-confident
and demanding, and harder working, than Wampar women. In the
same line of argument, many Wampar emphasised that the hierarchy in
such marital relationships is more clear-cut: whereas Wampar women
often talk back, make their own decisions and choices and have their
brothers and family to support them, yaner women are without their
male kin and in a more vulnerable position and tend to be mindful of
that fact. By the same token, they are also more likely to accept co-
wives in polygynous marriages. Some yaner women are more likely to
accept these vulnerable positions as some of them want to escape ex-
pectations and pressure in the place they come from for a better eco-
nomic future as described above.
Negative stereotypes about other ethnic groups are common among
the Wampar (Beer 2006a). In daily interactions and in the evaluation of
partners, however, individual qualities and potential are more impor-
tant and outweigh these negative images. Ideas about having a good fu-
ture life often depend in the younger generations not so much on physi-
cal attributes of the partner but more on chances to participate in an
imagined modern lifestyle, or chances to be able to develop autonomous
economic strategies and become more independent of the family of ori-
gin. Today Wampar women, for example, who marry non-Wampar
partners see the yaner in various respects as better potential husbands-
to-be than Wampar men: Ngaeng yaner could facilitate migration to
socio-economically interesting parts of the country (towns or certain
provinces), where they can live permanently or temporarily or in two
households, or if the non-Wampar partner stays on Wampar territory
the couple depends on her, her family and mainly her brothers. This
often gives the Wampar woman in an interethnic marriage more power
over her life as well as the future of the children in terms of education
and possible livelihoods. This outweighs the problem that sometimes
occurs of children of ngaeng yaner not being sure to be provided with
land on Wampar territory (see Bacalzo 2012, Bacalzo, Beer and Schwö-
rer 2014).
6. Interethnic Marriages Across Space and Time
Over the course of the 1990s, the Highlands Highway became in-
creasingly important to the economic life of the Wampar, who now of-
fer for sale to travellers everything from fruits, prepared food and beer
Social Reproduction and Ethnic Boundaries 17
Sociologus 64 (2014) 1
to kerosene and diesel. Prostitution has also become a source of income,
which has implications for the risk of HIV/ AIDS infections.11 The
Highlands Highway is the main link between the interior of the country
and the coast and the most important of the few roads in Papua New
Guinea. It has always been used to transport people and goods from the
highlands provinces to the coast, but since airfares have increased it
has also been used by large numbers of passengers travelling by public
motor vehicles. Bryant Allen estimates based on the official 2000 popu-
lation census that the Highlands Highway between Goroka (Eastern
Highlands) and Lae probably has ten times more village generated traf-
fic than any other road in Papua New Guinea, and that does not take
into account the trucks (Allen, pers. comm.). Although hold-ups are fre-
quent along the highway, they do not deter many people from travel-
ling.
Settlement patterns are closely connected to economic opportunities
and recent changes in Wampar residential preferences have had impli-
cations for the frequency with which yaner and Wampar meet as poten-
tial partners. In turn, the residential preferences of parties to intereth-
nic marriages facilitate encounters between members of different eth-
nic groups. Here the temporal and spatial positioning of actors acts to
(re-)structure Wampar social life in the process of reproducing it (Gid-
dens 2007), a process we describe in more detail in the following sec-
tions.
6.1 Changing Settlement Patterns
and Residence of Interethnic Couples
With the advent of colonialism, the Wampar were concentrated into
the main villages that still exist today. However, census data show that
over the last twenty years more and more families have moved out of
villages to live either in hamlets near gardens (to protect garden pro-
ducts against theft) or near the highway. The most important economic
activities for families take place outside the village, in distant gardens,
on chicken farms (which are built outside the village), around the air-
port or at markets along the highway.
The diversification in contemporary religious affiliation has contrib-
uted to the redistribution of the Wampar population and development
18 Bettina Beer and Julia H. Schroedter
Sociologus 64 (2014) 1
11 This has implications for interethnic marriages, as yaner women (more rarely
men) are blamed for having brought the disease into the Wampar area and spread
it by sexual relations (no matter if short-term, as co-wives, or inmarried women)
with ngaeng Wampar. On Wampar territory, the Highlands Highway combines
many conditions relevant to a consideration of the risk of HIV transmission (Beer
2008).
of a division between the Highlands Highway settlements and the still
Lutheran dominated villages. The congregations of the Seventh Day
Adventists (SDA), the Assembly of God (AOG), the Lutheran Renewal
and PNG Revival churches are located along the highway on Wampar
territory. The SDA church several years ago ‘bought’land near the
highway, from a Wampar businessman who claimed to have the right
to sell it and it has proved a constant source of conflict between SDA
adherents and Lutherans.12 Some of the non-Lutheran congregations
are run by yaner pastors who try to attract Wampar people as members.
Today all polygynously married Wampar men (including many of the
described unclear cases) settle along the Highlands Highway or in
towns outside Wampar territory, for church elders exclude them from
the Lutheran community and the general understanding is that they
should not live within the village. Some of them (or their wives) are
members of some of the new churches, which are less strident concern-
ing these forms of marriage.
Life near the highway offers commercial opportunities and gives ea-
sier access to town. On the other hand it has also some risks: traffic ha-
zards, social conflicts and the rates of criminality are much higher close
to the markets and the highway. For Wampar-Wampar couples the
costs often –but not always –outweigh the benefits, particularly if they
have fertile gardens in the bush they want to protect. For interethnic
couples the situation is different. Many of them work in town and settle
near the highway especially because ngaeng yaner do not own land and
have limited access to gardens from their in-laws. Thus the highway of-
fers them opportunities for small business not based on gardening such
as retailing of garden produce, cooked food, or goods bought in town.
This might explain the high percentage of 24.0% of marriages involving
ngaeng yaner settling at the highway (cf. table 5). Yet, the rate of inter-
marriages of Wampar men with afi yaner who set up households near
the Highlands Highway is even higher: it amounts to 34.8%. The con-
text of market settlements along the highway –one is called mix mar-
ket for all the intermarried couples living there –and the new congre-
gations not only attract interethnic couples, but they also increase the
likelihood of further meetings with yaner, either as they travel along
the Highlands Highway or they come to visit relatives already married
to Wampar and living there.
Social Reproduction and Ethnic Boundaries 19
Sociologus 64 (2014) 1
12 Some members of congregations who are allowed to work on Sunday (unlike
Lutherans) set up ‘Sunday markets’along the highway. This has created conflicts
with Wampar who themselves follow the rules of the Lutheran church according
to which the main market is closed on Sundays and after a death.
Table 5
Residence of Inter- and Intraethnic Couples
of Gabsongkek Wampar in 2009, Percent in Columns
Place of residence 2009 Type of marriage Total n
Intraethnic
(both
WamparI) Interethnic
Marriages of WamparI men
Village, Gabsongkeg proper and
divisions within 36.0% 18.2% 28.0% 116
Hamlets on Gabsongkeg territory 35.1 % 32.1 % 33.7 % 140
Along the Highlands Highway 20.6% 34.8% 27.0% 112
Other Wampar villages 2.2% 0.0% 1.2% 5
Lae (next town) 1.3% 3.7% 2.4% 10
Other parts of PNG 1.8% 6.4% 3.9% 16
Two places of residence 3.1% 4.8% 3.9% 16
Total 100.0% 100.0% 100.0% 415
Marriages of WamparI women
Village, Gabsongkeg proper and
divisions within 36.0% 10.3% 24.8% 100
Hamlets on Gabsongkeg territory 35.1 % 19.4 % 28.3 % 114
Along the Highlands Highway 20.6% 24.0% 22.1% 89
Other Wampar villages 2.2% 1.7% 2.0% 8
Lae (next town) 1.3% 6.9% 3.7% 15
Other parts of PNG 1.8% 32.0% 14.9% 60
Two places of residence 3.1% 5.7% 4.2% 17
Total 100.0% 100.0% 100.0% 403
Source: Ethnographic census data 1954 to 2009, men n=228 (intramarriage), n=187
(intermarriage); women n=228 (intramarriage), n=175 (intermarriage).
Over the last five years some Wampar groups have begun to lease or
sell plots of land to non-Wampar most of which tend, for obvious rea-
sons, to be near the highway. Although this practice is highly contro-
versial it has led to an increase in the number of yaner living on Wam-
par land, including families who are not related in any way to Wam-
par.13 This, like any increase in the number of co-resident yaner,
provides further opportunities for interethnic relations to develop, but
it also tends to blur social and territorial boundaries, which acts to dis-
20 Bettina Beer and Julia H. Schroedter
Sociologus 64 (2014) 1
13 A young Wampar founded a land-awareness theatre group, which travels to
different places to stage plays about the dangers of selling Wampar land to busi-
nessmen.
solve the connection between ‘Wamparness’and place. For many Wam-
par being Wampar was strongly related to their shared history in gaining
and holding the Markham Valley (Fischer 2013) and to their exclusive
right to those lands (Fischer 1975). For ordinary Wampar this has guar-
anteed a productive subsistence base for themselves and their descen-
dants and for migrants it has meant the perpetual possibility of a return
to security. Most Wampar still absolutely refuse the idea that land can
be ‘owned’in such a way that it can be sold. Indeed, some land that had
been ‘sold’has already been claimed back by relatives of the seller.14
Many afi Wampar married to yaner men live with their spouses in
towns in other parts of PNG (32.0%, cf. lower part of table 5).15 This
corresponds with the norm that women should move to the place of ori-
gin/ residence of their husbands, as they may pass on land rights to
their children. It also reflects the relatively high ethnic/ social status of
their yaner husbands, and their own qualifications as nurses, teachers,
or secretaries.
6.2 Opportunities to Meet Future Partners
The most important locations where future couples meet are on Gab-
songkeg territory (often along the highway), in the nearby town of Lae
and for the younger generation at boarding schools in different pro-
vinces of PNG. The occasions and contexts of crucial importance are
markets, school, work and shopping in town. Meetings in more remote
areas –in the mountains or distant islands –are rarer and happen
usually on the occasion of visiting relatives of intermarried Wampar,
or on journeys to other parts of PNG for church or sports activities
(cf. partner-choice in PNG towns, Rosi and Zimmer-Tamakoshi 1993).
Future configurations of Wampar social identity will be strongly af-
fected by the growing number of ‘mixed children’(miks pikinini), off-
spring of the increasing number of intermarried couples, negotiating
ethnic affiliation (Bacalzo 2011, 2012) and their partner choices. As we
showed many offspring of interethnic marriages (n=88; 54 females and
34 males) marry yaner as well (cf. figure 2). Wampar tend to differenti-
ate according to the gender of the in-marrying parent. If the father is
yaner, not only are they often ethnonymically tagged, but their clan af-
filiation is sometimes contested:
Social Reproduction and Ethnic Boundaries 21
Sociologus 64 (2014) 1
14 The conditions under which an individual can get legal title over communal
land are immensely complicated and non-transparent to most Wampar.
15 The percentage is probably higher because it is likely that women married in
distant towns have not been included in the census.
In such cases a common form of differentiation involves the way children of the
marriage are referred to: miks pikinini (mixed child), miks manki (mixed boy),
miks meri (mixed girl), or pikinini bilong ngaeng yaner (child of a non-Wampar
man); but ethnonymic identifiers are also stressed, as for instance, meri Tolai (a
Tolai girl /woman) or man Sepik (a boy / man from the Sepik).While children of
non-Wampar fathers acknowledge these terms they also put emphasis on their
connections through their Wampar mothers, and want to be recognized as chil-
dren of Wampar women. (Bacalzo 2012: 335)
For miks pikinini (especially girls) with a contested status who want
to stay on Wampar territory it would seem to be better –in terms of se-
curing their future close to their family of orientation –to marry a
Wampar, but this is often not the case as the high percentage of out-
marriages among miks pikini shows.
We analysed the place of origin of partners and parents of 88 inter-
married miks pikini (54 women and 34 men). 62 are married to a part-
ner who is not from the province of their yaner parent and 26 to a part-
ner who is from the yaner parent’s province. That means roughly 30%
of all intermarriages might have been facilitated by relatives of the
non-Wampar parent, which possibility is also apparent from the fre-
quency of visiting of those linked by the marriage. The rest of the inter-
ethnic marriages of miks pikinini might also be explained by the fact
that they grew up outside Wampar territory or near the Highlands
Highway, because (as described above) interethnic couples are more
likely to settle there, and the highway is an important area where in-
teractions with non-Wampar take place. Out of the 54 intermarried
miks meri 18 miks meri married a partner from the same region of their
yaner parent (10 fathers and 8 mothers), among the 34 miks manki 8
married an afi yaner from their yaner parent’s place of origin (5 fathers
and 3 mothers).
More complex transformations of Wampar identity and social rela-
tions appear to be underway in the recently developed context of the
likelihood of mining and possibly other large-scale projects. Social pro-
cesses concerned with negotiating inclusion in those groups entitled to
royalties and preferential employment have begun in earnest (Bacalzo,
Beer, and Schwörer 2014). As elsewhere in PNG (e.g. Bainton 2009, Gil-
berthorpe 2013, Guddemi 1997), the adaptation of ‘traditional’group-
ings (to meet perceived legal requirements and conserve within group
benefits) involves less permeable social boundaries. In general, it seems,
these processes of inclusion and exclusion have a tendency to introduce
inequality and marginalization along lines of age, gender, place, and
group identity (Ballard and Banks 2003).
These observations link up with discussions of the formation and on-
going maintainance of boundaries between ethnic groups within an-
22 Bettina Beer and Julia H. Schroedter
Sociologus 64 (2014) 1
thropology. Ethnicity has not disappeared in a globalized world but has
‘survived’and is sometimes even strengthened as a function of inter-
group relations. Its nature is changing with the different contexts of
mobilization such as transnational negotiations of indigeneity or like
among the Wampar as a resource by parts of the population, in their
strategies for internal control within their own class-stratified ethnic
group (cf. Jenkins 2008: 93 ff.). In this context (inter)marriage is of cen-
tral interest not only as a social practice in conflict with rules and
values but also as a result of intentions and agency by men and women.
7. Conclusion: Growth and/ or Dissolution
of an Ethnic Group?
Consider the ancestral Wampar population, in 1900, for example:
practically an endogamous population, but with the primary groups/
social identities (clans) interlinked by the dense ties produced by exo-
gamous unions and the bilateral kinship these produce; the transge-
nerational reproduction of the means of life, which principally relied
on land and human labour, was also directly implicated in relations
within and between clans but in ways that were also contingent upon
the individual’s kindred; economic practice, cultural forms, individual
life-trajectories and the quality of social networks tended towards
broad, normatively coherent patterns that sometimes evoke images of
the “crystalline”in the anthropologist (Lévi-Strauss 1976: 30). Being
Wampar, for most of the population, hardly ever posed a problem, al-
though relations between the primary sub-groups that comprise that
population were the heart of political life.
Our analysis, we hope, shows clearly that the patterns upon which so-
cial reproduction depended were sensitive to configurations of affinity
that post-colonial social life made possible. Gradually, through the ac-
cretion of the effects of marriage choices on the constitution of Wampar
as a group, ‘Wamparness’as an identity became a question. Now, with
60–70% of the population marrying out, the ethnicity of the Wampar
is felt to be at stake. Wampar children and miks pikinini identify with
one or both parents, with their friends in the village, later with school-
mates, and they find their future partners ever more often outside their
own “ethnic group”. Still, most of them stay connected with their rela-
tives in Gabsongkeg, who, for example, help with school fees and bride-
wealth, tend garden land for their future gardens and help maintain a
sense that home remains secure. Such a sense of security does not only
reflect a form of nostalgia or contemporary financial imperatives, for
claims to land that may turn out to be an immensely valuable source
Social Reproduction and Ethnic Boundaries 23
Sociologus 64 (2014) 1
of income depend upon those intra-Wampar identities. Accordingly,
everyone seeks to maintain or cement the Wampar identities they can
claim, no matter where they find themselves: visits are frequent, non-
residents are ready to contribute (if only with time) to communal pro-
jects, to attend celebrations and offer political support. Nevertheless,
descending generations build up new identifications, as residents of the
city of Lae, or Nadzab as products of a specific school, college or uni-
versity and as members of the friendship networks such institutions
produce.
Wampar articulate their default understanding of ‘Wamparness’and
the rights associated with it in the idioms of patriliny and virilocalty,
even though such norms are frequently honoured through exceptions.
Accordingly, how, in the face of contemporary marriage patterns,
‘Wamparness’can be maintained presents itself as a frequently dis-
cussed and important question, for it is connected to land rights and
the allocation of other limited resources. So, who counts as a Wampar
person is a difficult and delicate issue. It is, however, a question that
only poses itself case by case for each kingroup and is sensitive to the
particular configuration of loyalties within it. Therefore there can be
no general answer to this question. Increasing infertility combined with
high numbers of immigrants gives a new basis to older anxieties about
how contact with “outsiders”would weaken the strong Wampar –not
just individual bodies, but the whole population. The selling of land
further weakens the aspect of territorial belonging, and identification.
The temporal and spatial processes involved show that Wampar man-
age even under today’s historically grown constraints to maintain a dis-
tinct social identity which is based on a joint history, territorial claims
and social networks (cf. Beer 2012) which can be rather in- or exclusive
based on the specific situation and the current conditions.
Some of the other Wampar villages along the highway have devel-
oped a “joint policy”that prohibits the sale or lease of land. Time will
tell how that “policy”works. More generally, responses to the questions
posed by contemporary life are highly sensitive to the circumstance of
each politically autonomous unit. Within Gabsongkeg, different line-
ages have different strategies: some have a clear agreement not to sell
any land, some do it, and some are deeply divided. There is especially
high demand among businessmen for land on Gabsongkeg territory be-
cause of easy access to the airport and its proximity to town. Gabsong-
keg also has a particularly long history of settlement of unrelated non-
Wampar from the Watut and the Erap, ethnic groups from slightly
more remote and poorer areas, on their territory. However, these set-
tlers depended on the goodwill of the Gabsongkeg patron. With new,
well-off white collar workers from town buying or leasing land, the
24 Bettina Beer and Julia H. Schroedter
Sociologus 64 (2014) 1
situation is different. This also impacts on intermarriages. A wampar
woman married to a yaner can settle on Wampar land bought from lin-
eages other than their own, if her brothers –against the older rules –
are not willing to give her land to use with her partner.
Our argument that ethnic identity in PNG can greatly constrain mo-
bility, and impact on almost all aspects of the individual’s life chances,
as well as the reproduction of rules and resources has wider theoretical
implications: the institution of marriage and ethnic endogamy –impor-
tant for the maintenance of ethnic identities –as well as actual partner
choice based on intentions and agency are central for understanding
local and regional patterns of social reproduction and stratification not
only in PNG. The above mentioned discussions in anthropology about
ethnicity and the maintenance or dissolution of ethnic boundaries
should focus more on (inter)marriage to study internal differentiation
within socially and ethnically more and more stratified groups.
Ethnic boundaries, marital unions and kinship networks are rarely
sharp demarcations between social entities but rather gradual zones of
interaction. Giddens’emphasis on the interconnectedness of the tem-
poral and spatial positioning of actors and the reproduction and trans-
formation of social relations and identities is helpful in analysing pro-
cesses of boundary making/ dissolution. As ethnicity is gradual and
ethnic identities are defined by many day-to-day practices, values,
rules and resources, the developments among Wampar are difficult to
predict although aspects of the process of negotiating these identities
have become clearer through the analyses of interethnic marriages.
Whether future inequalities have very much to do with being Wam-
par remains an open question that depends upon how yaner are defined
and treated. Potentially, proposed capital investments can generate
vast intraethnic and interethnic differentials in individual, household
and group income. Currently, everyone, from landless immigrants who
depend on the patronage or goodwill of others to successful business
men and women who lease out land to yaner and have a permanent in-
come depends upon local relations inflected by residential and kinship
contingencies. These, in turn, are dimensions of the outcomes of prior
marriage strategies and the politics of land (which is still normatively
lineage-based). How current strategies and policies might be affected
by the advent of proposed large-scale capital projects remains to be
seen; certainly, local fields of social relations will be impacted and the
constitution of ‘the Wampar’as a social, political, administrative and
cultural group will remain a complex question.
Social Reproduction and Ethnic Boundaries 25
Sociologus 64 (2014) 1
References
Bacalzo, Doris 2011. Processes of Becoming: Transcultural Socialization and
Childhood among the Wampar. Tsantsa, 16, pp. 154 –158.
–2012. Transformations in Kinship, Land Rights and Social Boundaries among
the Wampar in Papua New Guinea and the Generative Agency of Children of
Interethnic Marriages. Childhood, 19 (3), pp. 332 –345.
Bacalzo, Doris, Beer, Bettina and Schwörer, Tobias, in press. Mining Narratives,
the Revival of “Clans”and Other Changes in Wampar Social Imaginaries: a
Case Study from Papua New Guinea. Special issue on “Narratives of Mining
in the Pacific”of Le Journal de la Société des Océanistes.
Bainton, Nicholas A. 2009. Keeping the Network out of View: Mining, Distinc-
tions and Exclusion in Melanesia. Oceania, 79 (1), pp. 18 –33.
Ballard, Chris and Banks, Glenn 2003. Resource wars: The anthropology of
mining. Annual Review of Anthropology, 32, pp. 287 –313.
Barker, John 1996. Village Inventions: Historical Variations Upon a Regional
Theme in Uiaku, Papua New Guinea. Oceania 66 (3), pp. 211 –229.
Beer, Bettina 2006a. Stonhet and Yelotop: Body Images, Physical Markers and
Definitions of Ethnic Boundaries in Papua New Guinea. Anthropological Fo-
rum, 16 (2), pp. 105–122.
–2006b. Interethnic Marriages: Changing Rules and Shifting Boundaries among
the Wampar of Papua New Guinea. In: Barbara Waldis and Reginald Byron,
eds. Migration and Marriage. Heterogamy and Homogamy in a Changing
World. Münster: Lit, pp. 20–39.
–2008. Buying Betel and Selling Sex: Contested Boundaries, Risk Milieus, and
Discourses about HIV/ AIDS in the Markham Valley, Papua New Guinea. In:
Leslie Butt and Richard Eves, eds. Making Sense of AIDS: Culture, Sexuality,
and Power in Melanesia. Honolulu: University of Hawai’i Press, pp. 97 –115.
–2010. Interethnische Beziehungen und Transkulturelle Verwandtschaft. In:
Edmute Alber, Bettina Beer, Julia Pauli and Michael Schnegg, eds. Ver-
wandtschaft Heute: Positionen, Ergebnisse und Perspektiven. Berlin: Reimer,
pp. 145–171.
–2012. “Afaro –ich verbinde”. Zur Herstellung transethnischer verwandtschaf-
tlicher Netzwerke am Beispiel der Wampar (Papua New Guinea). In: Markus
Gamper, Linda Reschke and Michael Schönhuth, eds. Knoten und Kanten 2.0.
Soziale Netzwerkanalyse in Medienforschung und Anthropologie. Bielefeld:
Transcript, pp. 285–310.
–in press. Cross-sex siblingship and marriage: Transformations of kinship rela-
tions among the Wampar, Papua New Guinea. Anthropologica.
Buss, David M. 1989. Sex differences in human mate preferences: Evolutionary
hypotheses tested in 37 cultures. Behavioral and Brain Sciences, 12, pp. 1 –49.
26 Bettina Beer and Julia H. Schroedter
Sociologus 64 (2014) 1
Buss, David M. and Angleitner, Alois 1989. Mate selection preferences in Ger-
many and the United States. Personality and Individual Differences, 10,
pp. 1269–1280.
Buss, David M., Shackelford, Todd K., Kirkpatrick, Lee A. and Larsen, Randy J.
2001. A half century of mate preferences: The cultural evolution of values.
Journal of Marriage and Family, 63, pp. 491–503.
Buunk, Bram P., Dijkstra, Pieternel, Kenrick, Douglas T. and Warntjes, Astrid
2001. Age preferences for mates as related to gender, own age, and involve-
ment level. Evolution & Human Behavior, 22 (4), pp. 241 –250.
Feingold, Alan 1990. Gender differences in effects of physical attractiveness on
romantic attraction: A comparison across five research paradigms. Journal of
Personality and Social Psychology, 59 (5), pp. 981 –993.
Fischer, Hans 1975. Gabsongkeg ’71: Verwandtschaft, Siedlung und Landbesitz
in einem Dorf in Neuguinea. München: Renner.
–1992. Weisse und Wilde. Erste Kontakte und Anfänge der Mission. Berlin: Rei-
mer.
–1997. Zensusaufnahmen –Das Beispiel Gabsongkeg. In: Walter Schulze, Hans
Fischer and Hartmut Lang, eds. Geburt und Tod: Ethnodemographische Pro-
bleme, Methoden und Ergebnisse. Berlin: Reimer, pp. 37 –91.
–2013. Woher wir kamen: Moderne Elemente zur Herkunftsgeschichte der
Wampar, Papua-Neuguinea. Sociologus, 63 (1 –2), pp. 125 –145.
–(ed) 1978. Wampar. Berichte über die alte Kultur eines Stammes in Papua
New Guinea. Bremen: Übersee-Museum Bremen.
Franzen, Axel and Hartmann, Josef 2001. Die Partnerwahl zwischen Wunsch
und Wirklichkeit: Eine empirische Studie zum Austausch von physischer At-
traktivität und sozialem Status. In: Thomas Klein, ed. Partnerwahl und Hei-
ratsmuster. Sozialstrukturelle Voraussetzungen der Liebe. Opladen: Leske +
Budrich, pp. 183–206.
Giddens, Anthony 2007. The Constitution of Society, Outline of the Theory of
Structuration, Cambridge: Polity Press. Kindle edition.
Gilberthorpe, Emma 2013. Community development in Ok Tedi, Papua New
Guinea: The role of anthropology in the extractive industries. Community De-
velopment Journal, 48 (3), pp. 466 –483.
Golub, Alex 2014. Leviathans at the gold mine. Creating inidgenous and corpo-
rate actors in Papua New Guinea. Durham, London: Duke Press.
Guddemi, Philip 1997. Continuities, contexts, complexities, and transforma-
tions: Local land concepts of a Sepik people affected by mining exploration.
Anthropological Forum, 7 (4), pp. 629 –648.
Jenkins, Richard 2008. Rethinking Ethnicity. (2nd edition). London, Thousand
Oaks, New Delhi, Singapore: Sage.
Kramp, Rita 1999. Familienplanung in Gabensis: Fertilitätswandel aus Ethno-
graphischer Sicht. Berlin: Reimer.
Social Reproduction and Ethnic Boundaries 27
Sociologus 64 (2014) 1
Lévi-Strauss, Claude 1976. Structural Anthropology, Volume 2. London: Pen-
guin Books.
LiPuma, Edward 2000. Encompassing Others: The Magic of Modernity in Mela-
nesia. Ann Arbor: The University of Michigan Press.
National Statistical Office 2001. Papua New Guinea 2000 Census: Preliminary
Figures. Port Moresby: National Statistical Office.
–2009. Papua New Guinea Demographic and Health Survey 2006 National Re-
port. Port Moresby: National Statistical Office.
–2013, 2011. National Population and Housing Census of Papua New Guinea.
Final Figures. Port Moresby: National Statistical Office.
Paliwala, Adam Blaxter 2012. Language in Papua New Guinea: the Value of
Census Data. Language & Linguistics in Melanesia, 30, pp. 1 –31.
Rosi, Pamela and Zimmer-Tamakoshi, Laura 1993. Love and Marriage among
the Educated Elite in Port Moresby. In: Richard A. Marksbury, ed. The Busi-
ness of Marriage. Transformations in Oceanic Matrimony. Pittsburgh, London:
University of Pittsburgh Press, pp. 157 –174.
Schoen, Robert and Wooldredge, John D. 1989. Marriage Choices in North Caro-
lina and Virginia, 1969–71 and 1979 –81. Journal of Marriage and the Family,
51 (2), pp. 465–481.
Schulze, Walter 1997. Die ethnodemographische Analyse der Zensusdaten von
Gabsongkeg. In: Walter Schulze, Hans Fischer and Hartmut Lang, eds. Geburt
und Tod. Ethnodemographische Probleme, Methoden und Ergebnisse. Berlin:
Reimer. pp. 92–189.
Schulze, Walter, Fischer, Hans and Lang, Hartmut (eds) 1997. Geburt und Tod.
Ethnodemographische Probleme, Methoden und Ergebnisse. Berlin: Reimer.
Schwarz, Sascha and Hassebrauck, Manfred 2012. Sex and Age Differences in
Mate-Selection Preferences. Human Nature, 23 (4), pp. 447 –466.
Vial, L. G. 1937. Some statistical aspects of population in the Morobe District,
New Guinea. Oceania 8, pp. 383–397.
28 Bettina Beer and Julia H. Schroedter
Sociologus 64 (2014) 1
